Provider Matrix¶

Supported Methods (Base compute)¶

Supported Methods (Block Storage)¶

Supported Methods (Key pair management)¶

Provider Matrix¶

Supported Methods¶

Provider Matrix¶

Supported Methods (Main)¶

Supported Methods (CDN)¶

Provider Matrix¶

Supported Methods¶

Compute¶

• Node - represents a cloud or virtual server.
• NodeSize - represents node hardware configuration. Usually this is amount of the available RAM, bandwidth, CPU speed and disk size. Most of the drivers also expose an hourly price (in dollars) for the Node of this size.
• NodeImage - represents an operating system image.
• NodeLocation - represents a physical location where a server can be.
• NodeState - represents a node state. Standard states are: running, rebooting, terminated, pending, stopped, suspended, paused, erro, unknown.

Key Pair Management¶

• KeyPair - represents an SSH key pair object.

Block Storage¶

• StorageVolume - represents a block storage volume
• VolumeSnapshot - represents a point in time snapshot of a StorageVolume

Example 1 - Logging output to standard error¶

If you want the output to be logged to the standard error (on Linux) you can set it to /dev/stderr:

LIBCLOUD_DEBUG=/dev/stderr python my_script.py

Example output:

# -------- begin 4431824872 request ----------
curl -i -X GET -H 'Host: s3.amazonaws.com' -H 'X-LC-Request-ID: 4431824872' -H 'Content-Length: 0' -H 'User-Agent: libcloud/0.6.0-beta1 (Amazon S3 (standard))' 'https://s3.amazonaws.com:443/?AWSAccessKeyId=foo&Signature=bar'
# -------- begin 4431824872:4431825232 response ----------
HTTP/1.1 200 OK
X-Amz-Id-2: 1234
Server: AmazonS3
Transfer-Encoding: chunked
X-Amz-Request-Id: FFFFFFFFFF
Date: Tue, 01 Nov 2011 22:29:11 GMT
Content-Type: application/xml

171
<?xml version="1.0" encoding="UTF-8"?>
<ListAllMyBucketsResult xmlns="http://s3.amazonaws.com/doc/2006-03-01/"><Owner><ID>sada8932dsa8d30i</ID><DisplayName>kami</DisplayName></Owner><Buckets><Bucket><Name>test34324323</Name><CreationDate>2011-11-01T22:17:23.000Z</CreationDate></Bucket></Buckets></ListAllMyBucketsResult>
0

# -------- end 4431824872:4431825232 response ----------

Example 2 - Making JSON / XML response human friendly¶

Non-formatted JSON response:

LIBCLOUD_DEBUG=/dev/stderr python my_script.py

# -------- begin 23125648:23160304 response ----------
HTTP/1.1 200 OK
Content-Length: 1572
X-Compute-Request-Id: req-79ab42d8-a959-44eb-8dec-bc9458b2f4b3
Server: nginx/1.4.7
Connection: keep-alive
Date: Sat, 06 Sep 2014 14:13:37 GMT
Content-Type: application/json

{"servers": [{"status": "ACTIVE", "updated": "2014-09-06T14:13:32Z", "hostId": "561d56de25c177c422278d7ca5f8b210118348040b12afbad06f278a", "addresses": {"internet-routable": [{"OS-EXT-IPS-MAC:mac_addr": "fa:16:3e:3f:c0:a1", "version": 4, "addr": "10.100.100.101", "OS-EXT-IPS:type": "fixed"}]}, "links": [{"href": "http://nova/v2/d3b31ebfd32744d19d848f3e9c351869/servers/deb35f96-be41-431e-b931-6e615ec720f4", "rel": "self"}, {"href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/servers/deb35f96-be41-431e-b931-6e615ec720f4", "rel": "bookmark"}], "key_name": null, "image": {"id": "e9537ddd-6579-4473-9898-d211ab90f6d3", "links": [{"href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/images/e9537ddd-6579-4473-9898-d211ab90f6d3", "rel": "bookmark"}]}, "OS-EXT-STS:task_state": null, "OS-EXT-STS:vm_state": "active", "OS-SRV-USG:launched_at": "2014-09-06T14:13:32.000000", "flavor": {"id": "90c2a137-611b-4dd2-9d65-d4a0b0858531", "links": [{"href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/flavors/90c2a137-611b-4dd2-9d65-d4a0b0858531", "rel": "bookmark"}]}, "id": "deb35f96-be41-431e-b931-6e615ec720f4", "security_groups": [{"name": "default"}], "OS-SRV-USG:terminated_at": null, "OS-EXT-AZ:availability_zone": "nova", "user_id": "06dda7c06aa246c88d7775d02bc119ac", "name": "test lc 2", "created": "2014-09-06T14:13:12Z", "tenant_id": "d3b31ebfd32744d19d848f3e9c351869", "OS-DCF:diskConfig": "MANUAL", "os-extended-volumes:volumes_attached": [], "accessIPv4": "", "accessIPv6": "", "progress": 0, "OS-EXT-STS:power_state": 1, "config_drive": "", "metadata": {}}]}
# -------- end 23125648:23160304 response ----------

Human friendly formatted JSON response:

LIBCLOUD_DEBUG=/dev/stderr LIBCLOUD_DEBUG_PRETTY_PRINT_RESPONSE=1 python my_script.py

# -------- begin 41102928:41133624 response ----------
HTTP/1.1 200 OK
Content-Length: 1572
X-Compute-Request-Id: req-3ce8b047-55cd-4e20-bfeb-b65619696aec
Server: nginx/1.4.7
Connection: keep-alive
Date: Sat, 06 Sep 2014 14:14:38 GMT
Content-Type: application/json

{
    "servers": [
        {
            "OS-DCF:diskConfig": "MANUAL",
            "OS-EXT-AZ:availability_zone": "nova",
            "OS-EXT-STS:power_state": 1,
            "OS-EXT-STS:task_state": null,
            "OS-EXT-STS:vm_state": "active",
            "OS-SRV-USG:launched_at": "2014-09-06T14:13:32.000000",
            "OS-SRV-USG:terminated_at": null,
            "accessIPv4": "",
            "accessIPv6": "",
            "addresses": {
                "internet-routable": [
                    {
                        "OS-EXT-IPS-MAC:mac_addr": "fa:16:3e:3f:c0:a1",
                        "OS-EXT-IPS:type": "fixed",
                        "addr": "10.100.100.101",
                        "version": 4
                    }
                ]
            },
            "config_drive": "",
            "created": "2014-09-06T14:13:12Z",
            "flavor": {
                "id": "90c2a137-611b-4dd2-9d65-d4a0b0858531",
                "links": [
                    {
                        "href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/flavors/90c2a137-611b-4dd2-9d65-d4a0b0858531",
                        "rel": "bookmark"
                    }
                ]
            },
            "hostId": "561d56de25c177c422278d7ca5f8b210118348040b12afbad06f278a",
            "id": "deb35f96-be41-431e-b931-6e615ec720f4",
            "image": {
                "id": "e9537ddd-6579-4473-9898-d211ab90f6d3",
                "links": [
                    {
                        "href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/images/e9537ddd-6579-4473-9898-d211ab90f6d3",
                        "rel": "bookmark"
                    }
                ]
            },
            "key_name": null,
            "links": [
                {
                    "href": "http://nova/v2/d3b31ebfd32744d19d848f3e9c351869/servers/deb35f96-be41-431e-b931-6e615ec720f4",
                    "rel": "self"
                },
                {
                    "href": "http://nova/d3b31ebfd32744d19d848f3e9c351869/servers/deb35f96-be41-431e-b931-6e615ec720f4",
                    "rel": "bookmark"
                }
            ],
            "metadata": {},
            "name": "test lc 2",
            "os-extended-volumes:volumes_attached": [],
            "progress": 0,
            "security_groups": [
                {
                    "name": "default"
                }
            ],
            "status": "ACTIVE",
            "tenant_id": "d3b31ebfd32744d19d848f3e9c351869",
            "updated": "2014-09-06T14:13:32Z",
            "user_id": "06dda7c06aa246c88d7775d02bc119ac"
        }
    ]
}
# -------- end 41102928:41133624 response ----------

Non-formatted XML response:

LIBCLOUD_DEBUG=/dev/stderr python my_script.py

# -------- begin 33145616:33126160 response ----------
HTTP/1.1 200 OK
X-Amzn-Requestid: e84f62d0-368e-11e4-820b-8bf013dc269e
Date: Sun, 07 Sep 2014 13:00:13 GMT
Content-Length: 457
Content-Type: text/xml

<?xml version="1.0"?>
<ListHostedZonesResponse xmlns="https://route53.amazonaws.com/doc/2012-02-29/"><HostedZones><HostedZone><Id>/hostedzone/Z14L0C73CHH1DN</Id><Name>example1.com.</Name><CallerReference>41747982-568E-0DFC-8C11-71C23757C740</CallerReference><Config><Comment>test</Comment></Config><ResourceRecordSetCount>9</ResourceRecordSetCount></HostedZone></HostedZones><IsTruncated>false</IsTruncated><MaxItems>100</MaxItems></ListHostedZonesResponse>
# -------- end 33145616:33126160 response ----------

Human friendly formatted XML response:

LIBCLOUD_DEBUG=/dev/stderr LIBCLOUD_DEBUG_PRETTY_PRINT_RESPONSE=1 python my_script.py

# -------- begin 19444496:19425040 response ----------
HTTP/1.1 200 OK
X-Amzn-Requestid: 01c02441-368f-11e4-b616-9b9bd7509a8f
Date: Sun, 07 Sep 2014 13:00:56 GMT
Content-Length: 457
Content-Type: text/xml

<?xml version="1.0" ?>
<ListHostedZonesResponse xmlns="https://route53.amazonaws.com/doc/2012-02-29/">
    <HostedZones>
        <HostedZone>
            <Id>/hostedzone/Z14L0C73CHH1DN</Id>
            <Name>example1.com.</Name>
            <CallerReference>41747982-568E-0DFC-8C11-71C23757C740</CallerReference>
            <Config>
                <Comment>test</Comment>
            </Config>
            <ResourceRecordSetCount>9</ResourceRecordSetCount>
        </HostedZone>
    </HostedZones>
    <IsTruncated>false</IsTruncated>
    <MaxItems>100</MaxItems>
</ListHostedZonesResponse>

# -------- end 19444496:19425040 response ----------

1. Using http_proxy environment variable¶

By setting http_proxy environment variable you can specify which proxy to use for all of the outgoing requests for a duration / life-time of the process or a script.

Without authentication:

http_proxy=http://<proxy hostname>:<proxy port> python my_script.py

With basic auth authentication:

http_proxy=http://<username>:<password>@<proxy hostname>:<proxy port> python my_script.py

2. Passing http_proxy argument to the connection class constructor¶

By passing http_proxy argument to the libcloud.common.base.Connection class constructor, you can specify which proxy to use for a particular connection.

from libcloud.compute.drivers.dreamhost import DreamhostConnection

PROXY_URL_NO_AUTH = 'http://<proxy hostname>:<proxy port>'
PROXY_URL_BASIC_AUTH = 'http://<user>:<pass>@<proxy hostname>:<proxy port>'

conn = DreamhostConnection(host='dreamhost.com', port=443,
                           timeout=None, proxy_url=PROXY_URL_NO_AUTH)

3. Calling set_http_proxy method¶

Calling set_http_proxy method allows you to specify which proxy to use for all the outgoing requests which follow set_http_proxy method call.

This method also allows you to use a different proxy for each request as shown in the example below.

from pprint import pprint

from libcloud.compute.types import Provider
from libcloud.compute.providers import get_driver

PROXY_URL_NO_AUTH_1 = 'http://<proxy hostname 1>:<proxy port 2>'
PROXY_URL_NO_AUTH_2 = 'http://<proxy hostname 1>:<proxy port 2>'
PROXY_URL_BASIC_AUTH = 'http://<user>:<pass>@<proxy hostname>:<proxy port>'

cls = get_driver(Provider.RACKSPACE)
driver = cls('username', 'api key', region='ord')

# Use proxy 1 for this request
driver.connection.set_http_proxy(proxy_url=PROXY_URL_NO_AUTH_1)
pprint(driver.list_nodes())

# Use proxy 2 for this request
driver.connection.set_http_proxy(proxy_url=PROXY_URL_NO_AUTH_2)
pprint(driver.list_nodes())

1. Import ordering¶

Organize the imports in the following order:

1. Standard library imports
2. Third-party library imports
3. Local library (Libcloud) imports

Each section should be separated with a blank line. For example:

import sys
import base64

import paramiko

from libcloud.compute.base import Node, NodeDriver
from libcloud.compute.providers import Provider

2. Function and method ordering¶

Functions in a module and methods on a class should be organized in the following order:

1. “Public” functions / methods
2. “Private” functions / methods (methods prefixed with an underscore)
3. “Internal” methods (methods prefixed and suffixed with a double underscore)

For example:

class Unicorn(object):
    def __init__(self, name='fluffy'):
        self._name = name

    def make_a_rainbow(self):
        pass

    def _get_rainbow_colors(self):
        pass

    def __eq__(self, other):
        return self.name == other.name

Methods on a driver class should be organized in the following order:

1. Methods which are part of the standard API
2. Extension methods
3. “Private” methods (methods prefixed with an underscore)
4. “Internal” methods (methods prefixed and suffixed with a double underscore)

Methods which perform a similar functionality should be grouped together and defined one after another.

For example:

class MyDriver(object):
    def __init__(self):
        pass

    def list_nodes(self):
        pass

    def list_images(self):
        pass

    def create_node(self):
        pass

    def reboot_node(self):
        pass

    def ex_create_image(self):
        pass

    def _to_nodes(self):
        pass

    def _to_node(self):
        pass

    def _to_images(self):
        pass

    def _to_image(self):
        pass

Methods should be ordered this way for the consistency reasons and to make reading and following the generated API documentation easier.

3. Prefer keyword over regular arguments¶

For better readability and understanding of the code, prefer keyword over regular arguments.

Good:

some_method(public_ips=public_ips, private_ips=private_ips)

Bad:

some_method(public_ips, private_ips)

4. Don’t abuse **kwargs¶

You should always explicitly declare arguments in a function or a method signature and only use **kwargs and *args respectively when there is a valid use case for it.

Using **kwargs in many contexts is against Python’s “explicit is better than implicit” mantra and makes it for a bad and a confusing API. On top of that, it makes many useful things such as programmatic API introspection hard or impossible.

A use case when it might be valid to use **kwargs is a decorator.

Good:

def my_method(self, name, description=None, public_ips=None):
    pass

Bad (please avoid):

def my_method(self, name, **kwargs):
    description = kwargs.get('description', None)
    public_ips = kwargs.get('public_ips', None)

5. When returning a dictionary, document its structure¶

Dynamic nature of Python can be very nice and useful, but if (ab)use it in a wrong way it can also make it hard for the API consumer to understand what is going on and what kind of values are being returned.

If you have a function or a method which returns a dictionary, make sure to explicitly document in the docstring which keys the returned dictionary contains.

6. Prefer to use “is not None” when checking if a variable is provided or defined¶

When checking if a variable is provided or defined, prefer to use is foo is not None instead of if foo.

If you use if foo approach, it’s easy to make a mistake when a valid value can also be falsy (e.g. a number 0).

For example:

class SomeClass(object):
    def some_method(self, domain=None):
        params = {}

        if domain is not None:
            params['Domain'] = domain

1. Start a discussion on the mailing list¶

If you are implementing a big feature or a change, start a discussion on the mailing list first.

2. Open a new issue on our issue tracker¶

Go to our issue tracker and open a new issue for your changes there. This issue will be used as an umbrella place for your changes. As such, it will be used to track progress and discuss implementation details.

3. Fork our Github repository¶

Fork our Github git repository. Your fork will be used to hold your changes.

4. Create a new branch for your changes¶

For example:

git checkout -b <jira_issue_id>_<change_name>

5. Make your changes¶

6. Write tests for your changes and make sure all the tests pass¶

Make sure that all the code you have added or modified has appropriate test coverage. Also make sure all the tests including the existing ones still pass.

For more information on how to write and run tests, please see Testing page.

7. Commit your changes¶

Make a single commit for your changes. If a corresponding JIRA ticket exists, make sure the commit message contains the ticket number.

For example:

git commit -a -m "[LIBCLOUD-123] Add a new compute driver for CloudStack based providers."

8. Open a pull request with your changes¶

Go to https://github.com/apache/libcloud/ and open a new pull request with your changes. Your pull request will appear at https://github.com/apache/libcloud/pulls.

Make sure the pull request name is prefixed with a JIRA ticket number, e.g. [LIBCLOUD-436] Improvements to DigitalOcean compute driver and that the pull request description contains link to the JIRA ticket.

9. Wait for the review¶

Wait for your changes to be reviewed and address any outstanding comments.

10. Squash the commits and generate the patch¶

Once the changes has been reviewed, all the outstanding issues have been addressed and the pull request has been +1’ed, close the pull request, squash the commits (if necessary) and generate a patch.

git format-patch --stdout trunk > patch_name.patch

Make sure to use git format-patch and not git diff so we can preserve the commit authorship.

Note #1: Before you generate the patch and squash the commits, make sure to synchronize your branch with the latest trunk (run git pull upstream trunk in your branch), otherwise we might have problems applying it cleanly.

Note #2: If you have never used rebase and squashed the commits before, you can find instructions on how to do that in the following guide: squashing commits with rebase.

11. Attach a final patch with your changes to the corresponding JIRA ticket¶

Attach the generated patch to the JIRA issue you have created earlier.

Note about Github¶

Github repository is a read-only mirror of the official Apache git repository (https://git-wip-us.apache.org/repos/asf/libcloud.git). This mirror script runs only a couple of times per day which means this mirror can be slightly out of date.

You are advised to add a separate remote for the official upstream repository:

git remote add upstream https://git-wip-us.apache.org/repos/asf/libcloud.git

Github read-only mirror is used only for pull requests and code review. Once a pull request has been reviewed, all the comments have been addresses and it’s ready to be merged, user who submitted the pull request must close the pull request, create a patch and attach it to the original JIRA ticket.

Syncing your git(hub) repository with an official upstream git repository¶

This section describes how to synchronize your git clone / Github fork with an official upstream repository.

It’s important that your repository is in-sync with the upstream one when you start working on a new branch and before you generate a final patch. If the repository is not in-sync, generated patch will be out of sync and we won’t be able to cleanly merge it into trunk.

To synchronize it, follow the steps below in your git clone:

1. Add upstream remote if you haven’t added it yet

git remote add upstream https://git-wip-us.apache.org/repos/asf/libcloud.git

2. Synchronize your trunk branch with an upstream one

git checkout trunk
git pull upstream trunk

3. Create a branch for your changes and start working on it

git checkout -b my_new_branch

4. Before generating a final patch which is to be attached to the JIRA ticket, make sure your repository and branch is still in-sync

git pull upstream trunk

5. Generate a patch which can be attached to the JIRA ticket

git format-patch --stdout remotes/upstream/trunk > patch_name.patch

Contributing Bigger Changes¶

If you are contributing a bigger change (e.g. large new feature or a new provider driver) you need to have signed Apache Individual Contributor License Agreement (ICLA) in order to have your patch accepted.

You can find more information on how to sign and file an ICLA on the Apache website.

When filling the form, leave field preferred Apache id(s) empty and in the notify project field, enter Libcloud.

Context Managers¶

Context managers aren’t available in Python 2.5 by default. If you want to use them make sure to put from __future__ import with_statement on top of the file where you use them.

Exception Handling¶

There is no unified way to handle exceptions and extract the exception object in Python 2.5 and Python 3.x. This means you need to use a sys.exc_info()[1] approach to extract the raised exception object.

For example:

try:
    some code
except Exception:
    e = sys.exc_info()[1]
    print e

Utility functions for cross-version compatibility¶

You can find a lot of utility functions which make code easier to work with Python 2.x and 3.x in libcloud.utils.py3 module.

You can find some more information on changes which are involved in making the code work with multiple versions on the following link - Lessons learned while porting Libcloud to Python 3

1. Subscribe to the public mailing lists¶

If you haven’t yet, subscribe to {dev,users,commits}@apache.libcloud.org mailing lists. Committs mailing list is especially important because all of the JIRA notification, Github Pull Request notifications and build notifications are sent there.

2. Subscribe to the private mailing list¶

Subscribe to private@libcloud.apache.org by sending an email to private-subscribe@libcloud.apache.org. Keep in mind that this list is private and your subscription needs to be approved by a moderator.

3. Create PyPi account¶

Creating a PyPi account by going to https://pypi.python.org/pypi?%3Aaction=register_form and send an email to private@libcloud.apache.org requesting to be added as a maintainer to apache-libcloud package. Make sure to select a strong and unique password (pwgen -s 48 1 is your friend).

After you have registered go to “Your details” page and populate PGP Key ID field with your PGP key ID.

1. Pre-release check list¶

• Make sure tests pass on all the supported Python versions (tox)
• Make sure CHANGES file is up to date
• Make sure __version__ string in libcloud/__init__.py is up to date
• Remove the tox directory with rm -rf .tox
• Remove the _secrets_ file with rm libcloud/test/secrets.py

2. Update JIRA¶

• Create a new JIRA version for the release in question (if one doesn’t exist yet)
• Close all the corresponding JIRA tickets and set Fix Version/s field to the current version
• Release the JIRA version

3. Creating release artifacts¶

We have a script that runs the required setup.py commands and then hashes and signs the files. You will need the latest version of pip and the wheel package. To run it:

cd dist
./release.sh -u <yourusername>@apache.org

-u argument will be used to find a key with a matching email address in your local GPG database.

This should result in a set of apache-libcloud-${VERSION}.{tar.bz2,tar.gz,zip,whl}{,asc,md5,sha1} files that are suitable to be uploaded for a release.

Copy the artifacts in another directory, unpack one of them and test it with tox.

4. Tagging a release¶

Tag the tentative release with a -tentative postfix.

git tag <version> <commit hash>

For example:

git tag v0.15.0-tentative 105b9610835f99704996d861d613c5a9a8b3f8b1

5. Upload the release artifacts and start a [VOTE] thread¶

Upload all release artifacts including the whl files to your people.apache.org space. Then start a [VOTE] thread on the dev@libcloud.apache.org mailing list.

Once the vote has passed tag the release with a new tag, removing the -tentative postfix. Upload the release artifacts to Apache servers and Pypi.

6. Uploading release artifacts to Apache servers¶

• Add release artifacts to the dist SVN repository at https://dist.apache.org/repos/dist/release/libcloud/

  It may take up to a day for the artifacts to be available on all the Apache mirrors, but they should be instantly available at http://www.apache.org/dist/libcloud/.

• If there is more than one old release in the repository, delete rest of the old release and only leave current and previous release there. Old releases are automatically archived and available at https://dist.apache.org/repos/dist/release/libcloud/.

7. Publishing package to PyPi¶

For consistency and security reasons packages are always uploaded to PyPi manually using the web interface and not using the setup.py upload command.

• Run python setup.py register command. This will register a new version on PyPi, but it won’t upload the actual release artifacts.
• Go to the PyPi release management page, find a new release and click on “files” link.
• Once you are there, upload all the release artifacts (.tar.bz2, .tar.gz, .zip, and .whl). For File Type select Source (except for .whl file where you should select Python Wheel) and for Python Version select Any (ie. pure Python). Make sure to also select and upload a PGP signature for each file (PGP signature (.asc) field).

Once all the files have been uploaded, the page should look similar to the screenshot below.

8. Verifying the release artifact check sums¶

To verify that nothing went wrong doing the release process, run the ./dist/verify_checksums.sh script.

This script downloads the release artifacts from the Apache and PyPi server and makes sure that the MD5 check sums of the both files match.

Usage:

./dist/verify_checksums.sh <version>

For example

./dist/verify_checksums.sh apache-libcloud-0.13.2

9. Updating doap_libcloud.rdf file¶

Add information about the new release to the doap_libcloud.rdf file in the root of the main code repository.

10. Updating website¶

Check out the website using SVN: svn co https://svn.apache.org/repos/asf/libcloud/site/trunk

• Update the front page (source/index.html file)
• Update “Downloads” page (source/downloads.md file)
• Add a blog entry in the _posts directory.

Build the site locally and make sure everything is correct. Check the README.md file.

11. Sending announcements¶

• Send a release announcement to {dev,users}@libcloud.apache.org. If it’s a major release also send it to announce@apache.org.
• Send a release announcement to Twitter and Google+

Note: If the release fixes a security vulnerability, you should also send information about a vulnerability to the following full disclosure mailing lists:

• full-disclosure@lists.grok.org.uk
• bugtraq@securityfocus.com

The email you send should also be signed using your GPG key. You can find an example of such announcement on the following url.

If needed, use Apache URL shortening service - http://s.apache.org/

Release voting thread template¶

This template should be used when starting a voting thread for a new release.

Subject:

[VOTE] Release Apache Libcloud <version>

Body:

This is a voting thread for Libcloud <version>.

<short description and highlights of this release>

Full changelog can be found at <link to the CHANGES file for this release>.

Release artifacts can be found at <link to your Apache space where a release
artifacts can be found>.

KEYS file can found at https://dist.apache.org/repos/dist/release/libcloud/KEYS

Please test the release and post your votes.

+/- 1
[  ]  Release Apache Libcloud <version>

Vote will be opened until <date, it should be at minimum today + 3 days> (or longer, if needed).

Thanks,
<name>

For example:

Subject:

[VOTE] Release Apache Libcloud 0.13.2

Body:

This is a voting thread for Libcloud 0.13.2.

This is another primarily a bug-fix release. Previous release included a fix for the Content-Length bug which didn't fully fix the original issue. It missed out "raw" requests which are fixed in this release (LIBCLOUD-396).

This bug could manifest itself while uploading a file with some of the storage providers.

Besides this bug fix, it includes a couple of other smaller bug fixes and changes. Full change log
can be found at https://git-wip-us.apache.org/repos/asf?p=libcloud.git;a=blob;f=CHANGES;h=b7747f777afdeb63bcacf496d1d034f1b3287c31;hb=c4b3daae946049652a500a8515929b4cbf14a6b4

Release artifacts can be found at http://people.apache.org/~tomaz/libcloud/.

Please test the release and post your votes.

+/- 1
[  ]  Release Apache Libcloud 0.13.2

Vote will be opened until September 18th, 2013 (or longer, if needed).

Thanks,
Tomaz

Release announcement¶

This template should be used when sending out a release announcement.

Subject:

[ANNOUNCE] Apache Libcloud 0.13.1 release

Body:

Libcloud is a Python library that abstracts away the differences among
multiple cloud provider APIs. It allows users to manage cloud services
(servers, storage, loadbalancers, DNS) offered by many different providers
through a single, unified and easy to use API.

We are pleased to announce the release of Libcloud <version>!

<short description of the release which should include release highlights>

Full change log can be found at <link to CHANGES file for this release>

Download

Libcloud <version> can be downloaded from http://libcloud.apache.org/downloads.html
or installed using pip:

pip install apache-libcloud

Upgrading

If you have installed Libcloud using pip you can also use it to upgrade it:

pip install --upgrade apache-libcloud

Upgrade notes

A page which describes backward incompatible or semi-incompatible
changes and how to preserve the old behavior when this is possible
can be found at http://libcloud.apache.org/upgrade-notes.html.

Documentation

API documentation can be found at http://libcloud.apache.org/apidocs/<version>/.

We also have a new Sphinx documentation which can be found at https://libcloud.apache.org/docs/.

Bugs / Issues

If you find any bug or issue, please report it on our issue tracker
<https://issues.apache.org/jira/browse/LIBCLOUD>.
Don't forget to attach an example and / or test which reproduces your problem.

Thanks

Thanks to everyone who contributed and made this release possible! Full list of
people who contributed to this release can be found in the CHANGES file
<link to the changes file for this release>.

For example:

Subject:

[ANNOUNCE] Apache Libcloud 0.13.1 release

Body:

Libcloud is a Python library that abstracts away the differences among
multiple cloud provider APIs. It allows users to manage cloud services
(servers, storage, loadbalancers, DNS) offered by many different providers
through a single, unified and easy to use API.

We are pleased to announce the release of Libcloud 0.13.1!

This is a bug-fix only release. Among some smaller bugs it also fixes
Content-Length regression which broke create and update operations in
the Bluebox Compute and Azure Storage driver (LIBCLOUD-362, LIBCLOUD-3901).

Full change log can be found at <https://git-wip-us.apache.org/repos/asf?p=libcloud.git;a=blob;f=CHANGES;h=ca90c84e296ca82e2206eb86ed7364c588aad503;hb=602b6a7a27dca6990a38eb887e1d6615826387d5>

Download

Libcloud 0.13.1 can be downloaded from http://libcloud.apache.org/downloads.html
or installed using pip:

pip install apache-libcloud

Upgrading

If you have installed Libcloud using pip you can also use it to upgrade it:

pip install --upgrade apache-libcloud

Upgrade notes

A page which describes backward incompatible or semi-incompatible
changes and how to preserve the old behavior when this is possible
can be found at http://libcloud.apache.org/upgrade-notes.html.

Documentation

API documentation can be found at http://libcloud.apache.org/apidocs/0.13.1/.

We also have a new Sphinx documentation which can be found at https://libcloud.apache.org/docs/.
Keep in mind though, that this documentation reflects state in trunk which
includes some backward incompatible changes which aren't present in 0.13.1.
All the examples in the documentation which only work with trunk are clearly marked with a note.

Bugs / Issues

If you find any bug or issue, please report it on our issue tracker
<https://issues.apache.org/jira/browse/LIBCLOUD>.
Don't forget to attach an example and / or test which reproduces your problem.

Thanks

Thanks to everyone who contributed and made this release possible! Full list of
people who contributed to this release can be found in the CHANGES file
<https://git-wip-us.apache.org/repos/asf?p=libcloud.git;a=blob;f=CHANGES;h=ca90c84e296ca82e2206eb86ed7364c588aad503;hb=602b6a7a27dca6990a38eb887e1d6615826387d5>.

General¶

• Update Rackspace AUTH_URL (LIBCLOUD-738) [Brian Curtin]
• Fix LIBCLOUD_DEBUG mode so it works on Python 3.x. [Tomaz Muraus]
• Fix Libcloud code so it doesn’t throw an exception if simplejson < 2.1.0 is installed. (LIBCLOUD-714, GITHUB-577) [Erik Johnson]
• Fix endpoint URL for DimensionData Asia Pacific region. (GITHUB-585) [Anthony Shaw]
• Document potential time drift issue which could cause authentication in the GCE drivers to fail. (GITHUB-571) [Michal Tekel]
• Update documentation for EC2 - make sure they reflect region changes from 0.14 release. (GITHUB-606) [James Guthrie]

Compute¶

• Fixed malformed XML requests with Dimension Data driver. (LIBCLOUD-760, GITHUB-610) [Anthony Shaw]
• Update list of scopes for Google Compute Engine driver. (GITHUB-607) [Otto Bretz]
• Allow user to filter VPC by project in the CloudStack driver by passing project argument to the ex_list_vps method. (GITHUB-516) [Syed Mushtaq Ahmed]
• Add volume management methods and other various improvements and fixes in the RunAbove driver. (GITHUB-561) [ZuluPro]
• Add support and update Dimension Data driver to use API v2.0 by default. (LIBCLOUD-736, GITHUB-564) [Anthony Shaw]
• Add new ex_virtual_network_name and ex_network_config argument to the create_node` method in the Azure driver. With those arguments user can now specify which virtual network to use. (GITHUB-569) [Jesaja Everling]
• Fix create_node method in the GCE driver calling inexistent method (ex_get_disk instead of ex_get_volume). (GITHUB-574) [Alex Poms]
• Allow user to pass proxy_url keyword argument to the VCloud driver constructor. (GITHUB-578) [Daniel Pool]
• Various fixes and improvements in the DimensionData driver (support for creating servers in MCP 1 and 2 data center, performance improvements in the location fetching, etc.). (GITHUB-587, GITHUB-593, LIBCLOUD-750, LIBCLOUD-753) [Anthony Shaw]
• Added ex_assign_public_ip argument to create_node in the EC2 driver. (GITHUB-590) [Kyle Long]
• Added ex_terminate_on_shutdown argument to create_node in the EC2 driver. (GITHUB-595) [Kyle Long]
• Various fixes and improvements in the ex_authorize_security_group_ingress in the CloudStack driver. (LIBCLOUD-749, GITHUB-580) [Lionel Schaub]
• Add pricing information for Softlayer. (LIBCLOUD-759, GITHUB-603) [David Wilson]

Storage¶

• Add support for sa-east-1 region to the Amazon S3 driver. (GITHUB-562) [Iuri de Silvio]
• Fix handling of binary data in Local storage driver on Python 3. Now the file which is to be written or read from is opened in the binary mode (b flag). (LIBCLOUD-725, GITHUB-568) [Torf]

Loadbalancer¶

• Add a new driver for DimensionData load-balancing service (http://cloud.dimensiondata.com/). (LIBCLOUD-737, GITHUB-567) [Anthony Shaw]

DNS¶

• Update Google Cloud DNS API from ‘v1beta1’ to ‘v1’ (GITHUB-583) [Misha Brukman]
• Add new driver for AuroraDNS service. (GITHUB-562, LIBCLOUD-735) [Wido den Hollander]
• Fix “_to_record” in the Route53 driver - make sure it doesn’t throw if the record TTL is not available. [Tomaz Muraus]
• Add new driver for WorldWideDNS service (http://www.worldwidedns.net/home.asp). (GITHUB-566, LIBCLOUD-732) [Alejandro Pereira]
• Add new driver for DNSimple service (https://dnsimple.com/). (GITHUB-575, GITHUB-604, LIBCLOUD-739) [Alejandro Pereira, Patrick Humpal]
• Add new driver for PointDNS service (https://pointhq.com). (GITHUB-576, GITHUB-591, LIBCLOUD-740) [Alejandro Pereira]
• Add new driver for Vultr DNS service (https://www.vultr.com). (GITHUB-579, GITHUB-596, LIBCLOUD-745) [Alejandro Pereira, Janez Troha]
• Add new driver for Liquidweb DNS service (http://www.liquidweb.com/). (GITHUB-581, LIBCLOUD-746) [Oltjano Terpollari, Alejandro Pereira]
• Add new driver for Zonomi DNS hosting service (http://zonomi.com/). (GITHUB-582, LIBCLOUD-747) [Oltjano Terpollari, Alejandro Pereira]
• Add new driver for Durable DNS service (https://durabledns.com/). (GITHUB-588, LIBCLOUD-748) [Oltjano Terpollari, Alejandro Pereira]

General¶

• Use native ssl.match_hostname functionality when running on Python >= 3.2 and only require backports.ssl_match_hostname dependency on Python versions < 3.2. [Tomaz Muraus]

• Add support for AWS Signature version 4.

  Note: Currently only GET HTTP method is supported. (GITHUB-444) [Gertjan Oude Lohuis]

• Fix a bug in the debug mode logging (LIBCLOUD_DEBUG). Logging to the debug file would throw an exception if the text contained non-ascii characters. [Tomaz Muraus]

• Fix a bug with connection code throwing an exception if a port was a unicode type and not a str or int. (GITHUB-533, LIBCLOUD-716) [Avi Weit]

• Update is_valid_ip_address function so it also works on Windows. (GITHUB-343, GITHUB-498, LIBCLOUD-601, LIBCLOUD-686) [Nicolas Fraison, Samuel Marks]

• Add support for retrying failed HTTP requests.

  Retrying is off by default and can be enabled by setting LIBCLOUD_RETRY_FAILED_HTTP_REQUESTS environment variable. (GITHUB-515, LIBCLOUD-360, LIBCLOUD-709)

• Fix a bug in consuming stdout and stderr strams in Paramiko SSH client. In some cases (like connecting to localhost via SSH), exit_status_ready gets set immediately even before the while loop to consume the streams kicks in. In those cases, we will not have consumed the streams at all. (GITHUB-558) [Lakshmi Kannan]

Compute¶

• Google Compute now supports paginated lists including filtering. (GITHUB-491) [Lee Verberne]

• OpenStackNodeSize objects now support optional, additional fields that are supported in OpenStack 2.1: ephemeral_disk, swap, extra. (GITHUB-488, LIBCLOUD-682) [Greg Hill]

• StorageVolume objects now have an attribute state that holds a state variable that is standardized state across drivers. Drivers that currently support the state attribute are OpenStack and EC2. StorageVolume objects returned by drivers that do not support the attribute will have a state of None. When a provider returns a state that is unknown to the driver, the state will be UNKNOWN. Please report such states. A couple of drivers already put state fields in the extra fields of StorageVolumes. These fields were kept for backwards-compatibility and for reference. (GITHUB-476) [Allard Hoeve]

• StorageVolume objects on EC2 and OpenStack now have a key called snapshot_id in their extra dicts containing the snapshot ID the volume was based on. (GITHUB-479) [Allard Hoeve]

• OpenStack driver: deprecated ex_create_snapshot and ex_delete_snapshot in favor of create_volume_snapshot and destroy_volume_snapshot. Updated base driver method create_storage_volume argument name to be optional. (GITHUB-478) [Allard Hoeve]

• Add support for creating volumes based on snapshots to EC2 and OS drivers. Also modify signature of base NodeDriver.create_volume to reflect the fact that all drivers expect a StorageSnapshot object as the snapshot argument. (GITHUB-467, LIBCLOUD-672) [Allard Hoeve]

• VolumeSnapshots now have a created attribute that is a datetime field showing the creation datetime of the snapshot. The field in VolumeSnapshot.extra containing the original string is maintained, so this is a backwards-compatible change. (GITHUB-473) [Allard Hoeve]

• Improve GCE create_node, make sure ex_get_disktype function (GITHUB-448) [Markos Gogoulos]

• GCE driver fix to handle unknown image projects (GITHUB-447) [Markos Gogoulos]

• Allow user to pass ex_blockdevicemappings argument to the create_node method in the OpenStack driver. (GITHUB-398, LIBCLOUD-637) [Allard Hoeve]

• Fix list_volume_snapshots method in the EC2 driver so it comforms to the base API. (LIBCLOUD-664, GITHUB-451) [Allard Hoeve]

• Add volumes_attached attibute to node.extra in the OpenStack driver. (LIBCLOUD-668, GITHUB-462) [Allard Hoeve]

• Add the following new methods to the Linode driver: ex_list_volumes, ex_create_volume, ex_destroy_volume. (LIBCLOUD-649, GITHUB-430) [Wojciech Wirkijowski]

• Add list_volume_snapshots method to the OpenStack driver. (LIBCLOUD-663, GITHUB-450) [Allard Hoeve]

• Add Site to Site VPN functionality to CloudStack driver. (GITHUB-465) [Avi Nanhkoesingh]

• Add affinity group support to CloudStack driver (LIBCLOUD-671, GITHUB-468) [Mateusz Korszun]

• Add a support for a new AWS Frankfurt, Germany region (eu-central-1) to the EC2 driver using AWS Signature v4. (GITHUB-444) [Gertjan Oude Lohuis, Tomaz Muraus]

• Allow Filtering in EC2 list_images() driver (GITHUB-456, LIBCLOUD-667) [Katriel Traum]

• Add ex_list_ip_forwarding_rules() to CloudStack driver (GITHUB-483) [Atsushi Sasaki]

• Add AURORA compute driver (LIBCLOUD-641, GITHUB-477) [Wido den Hollander]

• Update ex_describe_tags method in the EC2 driver and allow user to list tags for any supported resource. Previously you could only list tags for a node or a storage volume. (LIBCLOUD-676, GITHUB-482) [John Kinsella]

• Various improvements in the HostVirual driver (code refactoring, support for managing “packages”). (LIBCLOUD-670, GITHUB-472) [Dinesh Bhoopathy]

• Add support for DigitalOcean API v2.0 while maintaining support for the old API v2.0.

  Note: API v2.0 is now used by default. To use the old API v1.0, pass api_version='1.0' argument to the driver constructor. (GITHUB-442) [Andrew Starr-Bochicchio]

• Add new d4. instance types to the EC2 driver. Also update EC2 pricing data. (GITHUB-490) [Tomaz Muraus]

• Add new driver for Microsft Azure Virtual Machines service. (LIBCLOUD-556, GITHUB-305, GITHUB-499, GITHUB-538) [Michael Bennett, davidcrossland, Richard Conway, Matt Baldwin, Tomaz Muraus]

• Fix VPC lookup method in CloudStack driver (GITHUB-506) [Avi Nanhkoesingh]

• Add new driver for the Dimension Data provider based on the OpSource driver. (LIBCLOUD-698, GITHUB-507, LIBCLOUD-700, GITHUB-513) [Anthony Shaw]

• Add “virtualmachine_id” attribute to the CloudStackAddress class in the CloudStack driver. (LIBCLOUD-679, GITHUB-485) [Atsushi Sasaki]

• Allow user to pass filters via arguments to the ex_list_port_forwarding_rules in the CloudStack driver. (LIBCLOUD-678, GITHUB-484) [Atsushi Sasaki]

• Fix an issue with list_nodes in the CloudSigma driver throwing an exception if a node in the list had a static IP. (LIBCLOUD-707, GITHUB-514) [Chris O’Brien]

• Don’t throw an exception if a limit for a particular CloudStack resource is “Unlimited” and not a number. (GITHUB-512) [Syed Mushtaq Ahmed]

• Allow user to pass ex_config_drive argument to the create_node method in the OpenStack driver. (LIBCLOUD-356, GITHUB-330) [Ryan Parrish]

• Add new driver for Cloudwatt (https://www.cloudwatt.com/en/) provider. (GITHUB-338) [ZuluPro]

• Add new driver for Packet (https://www.packet.net/) provider. (LIBCLOUD-703, GITHUB-527) [Aaron Welch]

• Update Azure VM pricing information and add information for new D instance types. (GITHUB-528) [Michael Bennett]

• Add ex_get_node and ex_get_volume methods to CloudStack driver. (GITHUB-532) [ZuluPro]

• Update CloudSigma driver so the “unavailable” and “paused” node state is correctly mapped to “error” and “paused” respectively. (GITHUB-517) [Chris O’Brien]

• Add SSH key pair management methods to the Gandi driver. (GITHUB-534) [ZuluPro]

• Add ex_get_node and ex_get_volume methods to Gandi driver. (GITHUB-534) [ZuluPro]

• Add fault attribute to the extra dictionary of the Node instance returned by the OpenStack driver. (LIBCLOUD-730, GITHUB-557) [Nick Fox]

• Add new driver for Onapp IaaS platform. (LIBCLOUD-691, GITHUB-502) [Matthias Wiesner]

• Allow user to inject custom data / script into the Azure node by passing ex_custom_data argument to the create_node method. (LIBCLOUD-726, GITHUB-554) [David Wilson]

• Add ex_create_cloud_service and ex_destroy_cloud_service method to the Azure driver. (LIBCLOUD-724, GITHUB-551) [David Wilson]

• Add support for passing user data when creating a DigitalOcean node (ex_user_data argument). (LIBCLOUD-731, GITHUB-559) [David Wilson]

• Allow user to specify which arguments are passed to list_nodes method which is called inside wait_until_running by passing ex_list_nodes_kwargs argument to the wait_until_running method. (ex_user_data argument). (LIBCLOUD-723, GITHUB-548) [David Wilson]

• Allow user to pass ex_volume_type argument to the create_volume method in the OpennStack driver. (GITHUB-553) [Rico Echwald-Tijsen]

• Add new driver for RunAbove (https://www.runabove.com) provider. (GITHUB-550) [ZuluPro]

• Fix a bug with exception being throw inside the CloudStack driver when the provider returned no error message in the body. (GITHUB-555) [Konstantin Skaburskas]

• Various improvements in the DigitalOcean driver: - Increase page size to API maximum. - Add ex_create_attr kwarg to create_node method. - Update all the list_* methods to use paginated requests - Allow user to specify page size by passing ex_per_page argument to the

  constructor.

  (LIBCLOUD-717, GITHUB-537) [Javier Castillo II]

Storage¶

• Fix a bug with authentication in the OpenStack Swift driver. (GITHUB-492, LIBCLOUD-635) [Tom Fifield]
• Add AuroraObjects Storage Driver. (GITHUB-540, LIBCLOUD-719) [Wido den Hollander]

Loadbalancer¶

• Add a new driver for Softlayer load-balancing service (https://www.softlayer.com/load-balancing). (GITHUB-500, LIBCLOUD-688) [Avi Weit]

DNS¶

• Fix a bug when a ZoneDoesntExist exception was thrown when listing records for a zone which has no records in the HostVirtual driver. (GITHUB-460) [Vanč Levstik]
• Correctly handle MX records priority in the Route53 driver. (GITHUB-469) [Vanč Levstik]
• Allow user to create an A record which points directly to the domain zone name in the Route53 driver. (GITHUB-469) [Vanč Levstik]
• Fix delete_zone method in the HostVirtual driver. (GITHUB-461) [Vanč Levstik]
• Fix parsing of the record name in the HostVirtual driver. (GITHUB-461) [Vanč Levstik]
• Add new driver for DigitalOcean DNS service. (GITHUB-505) [Javier Castillo II]

General¶

• Use match_hostname function from backports.ssl_match_hostname package to verify the SSL certificate hostname instead of relying on our own logic. (GITHUB-374) [Alex Gaynor]

Compute¶

• GCE driver updated to include ex_stop_node() and ex_start_node() methods. (GITHUB-442) [Eric Johnson]
• GCE driver now raises ResourceNotFoundError when the specified image is not found in any image project. Previously, this would return None but now raises the not-found exception instead. This fixes a bug where returning None caused ex_delete_image to raise an AttributeError. (GITHUB-441) [Eric Johnson]
• GCE driver update to support JSON format Service Account files and a PY3 fix from Siim Põder for LIBCLOUD-627. (LIBCLOUD-627, LIBCLOUD-657, GITHUB-438) [Eric Johnson]
• GCE driver fixed for missing param on ex_add_access_config. (GITHUB-435) [Peter Mooshammer]
• GCE driver support for HTTP load-balancer resources. (LIBCLOUD-605, GITHUB-429) [Lee Verberne]
• GCE driver updated to make better use of GCEDiskTypes. (GITHUB-428) [Eric Johnson]
• GCE driver list_images() now returns all non-deprecated images by default. (LIBCLOUD-602, GITHUB-423) [Eric Johnson]
• Improve GCE API coverage for create_node(). (GITHUB-419) [Eric Johnson]
• GCE Licenses added to the GCE driver. (GITHUB-420) [Eric Johnson]
• GCE Projects support common instance metadata and usage export buckets. (GITHUB-409) [Eric Johnson]
• Improvements to TargetPool resource in GCE driver. (GITHUB-414) [Eric Johnson]
• Adding TargetInstances resource to GCE driver. (GITHUB-393) [Eric Johnson]
• Adding DiskTypes resource to GCE driver. (GITHUB-391) [Eric Johnson]
• Fix boot disk auto_delete in GCE driver. (GITHUB-412) [Igor Bogomazov]
• Add Routes to GCE driver. (GITHUB-410) [Eric Johnson]
• Add missing ubuntu-os-cloud images to the GCE driver. (LIBCLOUD-632, GITHUB-385) [Borja Martin]
• Add new us-east-2 and us-east-3 region to the Joyent driver. (GITHUB-386) [ZuluPro]
• Add missing t2. instance types to the us-west-1 region in the EC2 driver. (GITHUB-388) [Matt Lehman]
• Add option to expunge VM on destroy in CloudStack driver. (GITHUB-382) [Roeland Kuipers]
• Add extra attribute in list_images for CloudStack driver. (GITHUB-389) [Loic Lambiel]
• Add ex_security_group_ids argument to the create_node method in the EC2 driver. This way users can launch VPC nodes with security groups. (GITHUB-373) [Itxaka Serrano]
• Add description argument to GCE Network. (GITHUB-397) [Eric Johnson]
• GCE: Improve MachineType (size) coverage of GCE API. (GITHUB-396) [Eric Johnson]
• GCE: Improved Images coverage. (GITHUB-395) [Eric Johnson]
• GCE: Support for global IP addresses. (GITHUB-390, GITHUB-394) [Eric Johnson]
• GCE: Add missing snapshot attributes. (GITHUB-401) [Eric Johnson]
• AWS: Set proper disk size in c3.X instance types. (GITHUB-405) [Itxaka Serrano]
• Fix a bug with handling of the ex_keyname argument in the Softlayer driver. (GITHUB-416, LIBCLOUD-647) [Dustin Oberloh]
• Update CloudSigma region list (remove Las Vegas, NV region and add new San Jose, CA and Miami, FL region). (GITHUB-417) [Viktor Petersson]
• Add ex_get_node method to the Joyent driver. (GITHUB-421) [ZuluPro]
• Add support for placement group management to the EC2 driver. (GITHUB-418) [Mikhail Ovsyannikov]
• Add new tok02 region to the Softlayer driver. (GITHUB-436, LIBCLOUD-656) [Dustin Oberloh]
• Add new Honolulu, HI endpoint to the CloudSigma driver. (GITHUB-439) [Stephen D. Spencer]
• Fix a bug with config_drive attribute in the OpenStack driver. New versions of OpenStack now return a boolean and not a string. (GITHUB-433) [quilombo]
• Add support for Abiquo API v3.x, remove support for now obsolete API v2.x. (GITHUB-433, LIBCLOUD-652) [David Freedman]
• Allow rootdisksize parameter in create_node CloudStack driver (GITHUB-440, LIBCLOUD-658) [Loic Lambiel]

Storage¶

• Allow user to pass headers argument to the upload_object and upload_object_via_stream method.

  This way user can specify CORS headers with the drivers which support that. (GITHUB-403, GITHUB-404) [Peter Schmidt]

• Fix upload_object_via_stream so it works correctly under Python 3.x if user manually passes an iterator to the method.

  Also improve how reading a file in chunks works with drivers which support chunked encoding - always try to fill a chunk with CHUNK_SIZE bytes instead of directly streaming the chunk which iterator returns.

  Previously, if iterator returned 1 byte in one iteration, we would directly send this as a single chunk to the API. (GITHUB-408, LIBCLOUD-639) [Peter Schmidt]

Loadbalancer¶

• Updates to CloudStack driver. (GITHUB-434) [Jeroen de Korte]

DNS¶

• New driver for Softlayer DNS service. (GITHUB-413, LIBCLOUD-640) [Vanč Levstik]
• Fix a bug with ex_create_multi_value_record method in the Route53 driver only returning a single record. (GITHUB-431, LIBCLOUD-650) [Itxaka Serrano]

General¶

• Add new OpenStackIdentity_3_0_Connection class for working with OpenStack Identity (Keystone) service API v3. [Tomaz Muraus]
• Add support for prettifying JSON or XML response body which is printed to a file like object when using LIBCLOUD_DEBUG environment variable. This option can be enabled by setting LIBCLOUD_DEBUG_PRETTY_PRINT_RESPONSE environment variable. [Tomaz Muraus]
• Add support for using an HTTP proxy for outgoing HTTP and HTTPS requests. [Tomaz Muraus, Philip Kershaw]